The effect of pertussis toxin on opioid antinociception was studied in rats. Intrathecal injection of a single dose of pertussis toxin reduced the antinociceptive effect of PL017, a highly selective mu-opioid agonist, in a dose- and time-dependent manner. The maximal effective dose of pertussis toxin was 1 microgram, and the maximal effect was seen at day 3. The effect of the toxin lasted for 2 weeks, and the antinociceptive response recovered partially at the third week. The dose-response curves of the antinociceptive effect of PL017 were shifted to the right with increasing doses of pertussis toxin. Three days after pertussis injection, receptor-binding activity of membranes in the lumbosacral segment of spinal cords decreased to 70% of control as assayed by 125I-FK33824, a highly selective mu-receptor agonist. In experiments using [3H]naloxone as the radiolabeled ligand, displacement curves of FK33824 were shifted to the right after pertussis toxin treatment. The shift also was dose and time dependent. Scatchard analysis of binding data showed that, after pertussis toxin treatment, there was no significant change in the total number of binding sites, but a class of low-affinity binding sites appeared in addition to the high-affinity sites. When spinal membranes were washed in Na+ (100 mM) and guanosine diphosphate (100 microM) and binding was assayed in the presence of Mg2+ (5 mM), all the mu-receptors were in the high-affinity state in control membranes. After the pertussis toxin treatment, the ratio of low-affinity sites to high-affinity sites increased.(ABSTRACT TRUNCATED AT 250 WORDS)